Sparse Autoencoder-based Multi-head Deep Neural Networks for Machinery Fault Diagnostics with Detection of Novelties

Chinese Journal of Mechanical Engineering

Table 2 Comparison of performance of novelty detection

	The proposed method	One-class SVM fed by multi-head DNN features	Isolation forest fed by multi-head DNN features
\({\varvec{T}}{\varvec{P}}{\varvec{R}}\)(%)	88.57	97.69	61.32
\({\varvec{T}}{\varvec{N}}{\varvec{R}}\)(%)	98.53	4.14	100
\({{\varvec{F}}}_{1}\) score	0.93	0.65	0.76